PATCH: PR driver/28437: [4.2 Regression] multiple fno-builtin-* flags broken

[Steven Bosscher]

On Tuesday 25 July 2006 01:00, H. J. Lu wrote:
> prune_options should skip joined switches since switches with the same
> prefix have the same cl_option entry. They are different switchs.

A much better approach would be to revert the patch that caused this
whole mess (http://gcc.gnu.org/ml/gcc-patches/2006-04/msg01013.html).

If you're going to patch the driver, you're going to have to make it
understand the options it is pruning (and you should definitely clean
up that horrible "cont:" label garbage in prune_options!).

My $0.02

Gr.
Steven